https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675469#M23626 <P>Do you want impute missing value by EM algorithm ?</P> <P><a href="https://communities.sas.com/t5/user/viewprofilepage/user-id/13684">@Rick_SAS</a>&nbsp; wrote a blog about it recently .</P> Sun, 09 Aug 2020 11:25:28 GMT Ksharp 2020-08-09T11:25:28Z https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675450#M23621 <P>Hello,</P><P>&nbsp;</P><P>I am running a logistic regression with four predictor variables, 1 outcome variable, and 6 covariates.&nbsp; My N is around 2000 participants.</P><P>&nbsp;</P><P>For one of my four predictor variables, there are around 1000 people missing the question.&nbsp; The question asked was very sensitive and it is normal for people to skip it on surveys per other papers. The situation is similar for one of my covariates, where a question on a sensitive topic is also skipped by a substantial number of people.&nbsp;</P><P>&nbsp;</P><P>Right now when I run it, my data set ends up cut in half.&nbsp; My major professor will not let me do multiple imputations and wants me to do Full Information Maximum Likelihood to account for the missing data.</P><P>&nbsp;</P><P>Does anyone have any advice on if I can do this in SAS?&nbsp;</P><P>&nbsp;</P><P>Thank you!</P> Sun, 09 Aug 2020 02:16:50 GMT https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675450#M23621 klongway 2020-08-09T02:16:50Z https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675463#M23624 From what you describe, the people are not "missing" the question, they are deliberately choosing not to answer it. This is information that might be useful for your model.<BR /><BR />You can try recoding this variable to "not answered" (or whatever) then run the logistic regression. Sun, 09 Aug 2020 08:25:20 GMT https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675463#M23624 Norman21 2020-08-09T08:25:20Z https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675467#M23625 <P>Some people will fill in the mean of the data in place of the missings and also create a new variable which is 0 if the original variable was not missing and 1 if the original variable was missing. Then you put the original variable and the new variable into the model. This essentially predicts the observations with the original variable missing at a constant offset from the observations when the original variable was not missing.</P> Sun, 09 Aug 2020 11:05:48 GMT https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675467#M23625 PaigeMiller 2020-08-09T11:05:48Z https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675469#M23626 <P>Do you want impute missing value by EM algorithm ?</P> <P><a href="https://communities.sas.com/t5/user/viewprofilepage/user-id/13684">@Rick_SAS</a>&nbsp; wrote a blog about it recently .</P> Sun, 09 Aug 2020 11:25:28 GMT https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675469#M23626 Ksharp 2020-08-09T11:25:28Z https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675472#M23627 <P><A href="https://go.documentation.sas.com/?docsetId=statug&amp;docsetTarget=statug_calis_examples36.htm&amp;docsetVersion=14.3&amp;locale=en" target="_self">The SAS documentation has an example</A> of how to compute FIML by using structural equation modeling with PROC CALIS.</P> Sun, 09 Aug 2020 11:37:46 GMT https://communities.sas.com/t5/New-SAS-User/Struggling-with-Missingness-in-a-data-set/m-p/675472#M23627 Rick_SAS 2020-08-09T11:37:46Z